Installing the appropriate non-IOMMU DMA ops in arm_iommu_detch_device()
serves the case where IOMMU-aware drivers choose to control their own
mapping but still make DMA API calls, however it also affects the case
when the arch code itself tears down the mapping upon driver unbinding,
where the ops now get left in place and can inhibit arch_setup_dma_ops()
on subsequent re-probe attempts.
Fix the latter case by making sure that arch_teardown_dma_ops() cleans
up whenever the ops were automatically installed by its counterpart.
Reported-by: Tobias Jakobi <[email protected]>
Reported-by: Marek Szyprowski <[email protected]>
Fixes: 1874619a7df4 "ARM: dma-mapping: Set proper DMA ops in arm_iommu_detach_device()"
Tested-by: Tobias Jakobi <[email protected]>
Signed-off-by: Robin Murphy <[email protected]>
---
Sorry for the delay - there was a giant email infrastructure cock-up just
at the point I wanted to go back through my archive and double-check the
discussion around the original commit...
Robin.
arch/arm/mm/dma-mapping.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index f1e2922e447c..1e3e08a1c456 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -2390,4 +2390,6 @@ void arch_teardown_dma_ops(struct device *dev)
return;
arm_teardown_iommu_dma_ops(dev);
+ /* Let arch_setup_dma_ops() start again from scratch upon re-probe */
+ set_dma_ops(dev, NULL);
}
--
2.20.1.dirty
On Mon, Jan 21, 2019 at 02:52:16PM +0000, Robin Murphy wrote:
> Installing the appropriate non-IOMMU DMA ops in arm_iommu_detch_device()
> serves the case where IOMMU-aware drivers choose to control their own
> mapping but still make DMA API calls, however it also affects the case
> when the arch code itself tears down the mapping upon driver unbinding,
> where the ops now get left in place and can inhibit arch_setup_dma_ops()
> on subsequent re-probe attempts.
>
> Fix the latter case by making sure that arch_teardown_dma_ops() cleans
> up whenever the ops were automatically installed by its counterpart.
>
> Reported-by: Tobias Jakobi <[email protected]>
> Reported-by: Marek Szyprowski <[email protected]>
> Fixes: 1874619a7df4 "ARM: dma-mapping: Set proper DMA ops in arm_iommu_detach_device()"
> Tested-by: Tobias Jakobi <[email protected]>
> Signed-off-by: Robin Murphy <[email protected]>
> ---
>
> Sorry for the delay - there was a giant email infrastructure cock-up just
> at the point I wanted to go back through my archive and double-check the
> discussion around the original commit...
>
> Robin.
>
> arch/arm/mm/dma-mapping.c | 2 ++
> 1 file changed, 2 insertions(+)
I had also tested your draft on Tegra last week and this looks
identical, so:
Tested-by: Thierry Reding <[email protected]>
Hi Robin,
On 2019-01-21 15:52, Robin Murphy wrote:
> Installing the appropriate non-IOMMU DMA ops in arm_iommu_detch_device()
> serves the case where IOMMU-aware drivers choose to control their own
> mapping but still make DMA API calls, however it also affects the case
> when the arch code itself tears down the mapping upon driver unbinding,
> where the ops now get left in place and can inhibit arch_setup_dma_ops()
> on subsequent re-probe attempts.
>
> Fix the latter case by making sure that arch_teardown_dma_ops() cleans
> up whenever the ops were automatically installed by its counterpart.
>
> Reported-by: Tobias Jakobi <[email protected]>
> Reported-by: Marek Szyprowski <[email protected]>
> Fixes: 1874619a7df4 "ARM: dma-mapping: Set proper DMA ops in arm_iommu_detach_device()"
> Tested-by: Tobias Jakobi <[email protected]>
> Signed-off-by: Robin Murphy <[email protected]>
> ---
>
> Sorry for the delay - there was a giant email infrastructure cock-up just
> at the point I wanted to go back through my archive and double-check the
> discussion around the original commit...
No problem, could you also upload it to rmk's patch tracking system?
IMHO rmk's tree will be the best place to handle this fix.
> Robin.
>
> arch/arm/mm/dma-mapping.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
> index f1e2922e447c..1e3e08a1c456 100644
> --- a/arch/arm/mm/dma-mapping.c
> +++ b/arch/arm/mm/dma-mapping.c
> @@ -2390,4 +2390,6 @@ void arch_teardown_dma_ops(struct device *dev)
> return;
>
> arm_teardown_iommu_dma_ops(dev);
> + /* Let arch_setup_dma_ops() start again from scratch upon re-probe */
> + set_dma_ops(dev, NULL);
> }
Best regards
--
Marek Szyprowski, PhD
Samsung R&D Institute Poland